Currently, a control system with relatively high security and stability requirement is equipped with more than one servers or controllers. The servers or controllers in the system mostly employ a standby redundancy, or restart the servers or controllers that abnormally quit to solve the security and stability problem in the system. However, in the standby redundancy mode, if both the master machine and the standby machine fail, the system will be out of control, leading to poor security and stability of the system. The approach to restart a controller that abnormally quits will cause data loss of the control master machine and the control node after an abnormal system quit resulting in the system unable to operate normally after the system restarts.